Модель прогнозирования живого веса с помощью глубокой регрессии RGB-D изображений

Алексей Николаевич Ручай

Аннотация


Прогнозирование живого веса помогает контролировать здоровье животных, эффективно проводить генетическую селекцию и определять оптимальное время убоя. На крупных фермах для измерения живого веса используются точные и дорогостоящие промышленные весы. Взвешивание животного из-за стресса ведет к потере его веса и продуктивности на 5-10%. Однако, перспективной альтернативой является оценка живого веса с помощью морфометрических измерений животного, а затем применение уравнений регрессии, связывающих такие измерения с живым весом. Ручные измерения животных с помощью рулетки отнимают много времени и вызывают стресс у животных. Поэтому в настоящее время для бесконтактных морфометрических измерений все чаще используются технологии компьютерного зрения. В статье предлагается новая модель для прогнозирования живого веса на основе регрессии изображений с использованием методов глубокого обучения. Для регрессии изображений использовались RGB изображения и карты глубины вид сбоку для прогнозирования живого веса крупного рогатого скота. Показано, что на реальных наборах данных предложенная модель достигает точности измерения веса с ошибкой MAE 35.5 и MAPE 8.4 на тестовом наборе данных.

Ключевые слова


регрессия изображений; прогнозирование живого веса; глубокое обучение

Полный текст:

PDF

Литература


Wang Z., Shadpour S., Chan E., et al. ASAS-NANP SYMPOSIUM: Applications of machine learning for livestock body weight prediction from digital images. Journal of Animal Science. 2021. Vol. 99, no. 2. DOI: 10.1093/jas/skab022.

Ruchay A., Kober V., Dorofeev K., et al. Accurate body measurement of live cattle using three depth cameras and non-rigid 3-D shape recovery. Computers and Electronics in Agriculture. 2020. Vol. 179. P. 105821. DOI: 10.1016/j.compag.2020.105821.

Kuzuhara Y., Kawamura K., Yoshitoshi R., et al. A preliminarily study for predicting body weight and milk properties in lactating Holstein cows using a three-dimensional camera system. Computers and Electronics in Agriculture. 2015. Vol. 111. P. 186–193. DOI: 10.1016/j.compag.2014.12.020.

Sawanon S., Boonsaen P., Innurak P. Body Measurements of Male Kamphaeng Saen Beef Cattle as Parameters for Estimation of Live Weight. Kasetsart Journal - Natural Science. 2011. Vol. 45, no. 3. P. 428–434.

Wangchuk K., Wangdi J., Mindu M. Comparison and reliability of techniques to estimate live cattle body weight. Journal of Applied Animal Research. 2017. Vol. 46. P. 4. DOI: 10.1080/09712119.2017.1302876.

Vanvanhossou F., Diogo R., Dossa L. Estimation of live bodyweight from linear body measurements and body condition score in the West African Savannah Shorthorn Cattle in North-West Benin. Cogent Food And Agriculture. 2018. Vol. 4, no. 1. P. 1549767. DOI: 10.1080/23311932.2018.1549767.

Huma Z., Iqbal F. Predicting the body weight of Balochi sheep using a machine learning approach. Turkish journal of veterinary and animal sciences. 2019. Vol. 43, no. 4. P. 500–506. DOI: 10.3906/vet-1812-23.

Hempstalk K., Mcparland S., Berry D. Machine learning algorithms for the prediction of conception success to a given insemination in lactating dairy cows. Journal of dairy science. 2015. Vol. 98, no. 8. P. 5262–5273. DOI: 10.3168/jds.2014-8984.

Miller G.A., Hyslop J.J., Barclay D., et al. Using 3D Imaging and Machine Learning to Predict Liveweight and Carcass Characteristics of Live Finishing Beef Cattle. Frontiers in Sustainable Food Systems. 2019. Vol. 3. P. 30. DOI: 10.3389/fsufs.2019.00030.

Milosevic B., Ciric S., Lalic N., et al. Machine learning application in growth and health prediction of broiler chickens. World’s Poultry Science Journal. 2019. Vol. 75. P. 401–410. DOI: 10.1017/S0043933919000254.

Weber V., Weber F., Gomes R., et al. Prediction of Girolando cattle weight by means of body measurements extracted from images. Revista Brasileira de Zootecnia. 2020. Mar. Vol. 49. DOI: 10.37496/rbz4920190110.

Tasdemir S., Urkmez A., Inal S. Determination of body measurements on the Holstein cows using digital image analysis and estimation of live weight with regression analysis. Computers and Electronics in Agriculture. 2011. Vol. 76, no. 2. P. 189–197. DOI: 10.1016/j.compag.2011.02.001.

Pezzuolo A., Milani V., Zhu D., et al. On-Barn Pig Weight Estimation Based on Body Measurements by Structure-from-Motion (SfM). Sensors. 2018. Vol. 18, no. 11. Article 3603. DOI: 10.3390/s18113603.

Song X., Bokkers E., Tol P. van der, et al. Automated body weight prediction of dairy cows using 3-dimensional vision. Journal of Dairy Science. 2018. Vol. 101, no. 5. P. 4448–4459. DOI: 10.3168/jds.2017-13094.

Ranganathan H., Venkateswara H., Chakraborty S., Panchanathan S. Deep Active Learning for Image Regression. Deep Learning Applications. Singapore: Springer Singapore, 2020. P. 113–135. DOI: 10.1007/978-981-15-1816-4_7.

Bezsonov O., Lebediev O., Lebediev V., et al. Breed Recognition and Estimation of Live Weight of Cattle Based on Methods of Machine Learning and Computer Vision. Eastern-European Journal of Enterprise Technologies. 2021. Vol. 6/9, no. 114. P. 64–74. DOI: 10.15587/1729-4061.2021.247648.

Ruchay A., Dorofeev K., Kalschikov V., et al. Live weight prediction of cattle using deep image regression. 2021 IEEE International Workshop on Metrology for Agriculture and Forestry (MetroAgriFor). 2021. P. 32–36. DOI: 10.1109/MetroAgriFor52389.2021.9628547.

Ruchay A., Dorofeev K., Kober A., et al. Accuracy analysis of 3D object shape recovery using depth filtering algorithms. Applications of Digital Image Processing XLI. Vol. 10752. SPIE, 2018. P. 1075221–10. DOI: 10.1117/12.2319907.

Ruchay A., Kolpakov V., Kosyan D., et al. Genome-Wide Associative Study of Phenotypic Parameters of the 3D Body Model of Aberdeen Angus Cattle with Multiple Depth Cameras. Animals. 2022. Vol. 12, no. 16. Article 2128. DOI: 10.3390/ani12162128.

Lu J., Guo H., Du A., et al. 2-D/3-D fusion-based robust pose normalisation of 3-D livestock from multiple RGB-D cameras. Biosystems Engineering. 2021. Vol. 223. P. 129–141. DOI: 10.1016/j.biosystemseng.2021.12.013.

Bochkovskiy A., Wang C.-Y., Liao H.-Y.M. YOLOv4: Optimal Speed and Accuracy of Object Detection. CoRR. 2020. Vol. abs/2004.10934. arXiv: 2004.10934. URL: https://arxiv.org/abs/2004.10934.

Hu Y., Luo X., Gao Z., et al. Curve Skeleton Extraction from Incomplete Point Clouds of Livestock and Its Application in Posture Evaluation. Agriculture. 2022. Vol. 12, no. 7. Article 998. DOI: 10.3390/agriculture12070998.

Ruchay A., Kober V. Clustered impulse noise removal from color images with spatially connected rank filtering. Applications of Digital Image Processing XXXIX. Vol. 9971. SPIE, 2016. 99712Y–10. DOI: 10.1117/12.2236785.

Ruchay A., Kober V. Removal of impulse noise clusters from color images with local order statistics. Applications of Digital Image Processing XL. Vol. 10396. SPIE, 2017. P. 1039626–10. DOI: 10.1117/12.2272718.

Ruchay A., Kober V. Impulsive noise removal from color video with morphological filtering. Applications of Digital Image Processing XL. Vol. 10396. SPIE, 2017. P. 1039627–9. DOI: 10.1117/12.2272719.

Ruchay A., Kober V. Impulsive Noise Removal from Color Images with Morphological Filtering. Analysis of Images, Social Networks and Texts. Vol. 10716. Cham: Springer International Publishing, 2018. P. 280–291. Lecture Notes in Computer Science. DOI: 10.1007/978-3-319-73013-4_26.

Ruchay A., Dorofeev K., Kalschikov V. A novel switching bilateral filtering algorithm for depth map. Computer Optics. 2019. Vol. 43, no. 6. P. 1001–1007. DOI: 10.18287/2412-6179-2019-43-6-1001-1007.

Ruchay A.N., Dorofeev K.A., Kalschikov V.. Accuracy analysis of 3D object reconstruction using point cloud filtering algorithms. CEUR Workshop Proceedings. 2019. Vol. 2391. P. 169–174. DOI: 10.18287/1613-0073-2019-2391-169-174.

Rusu R.B., Cousins S. 3D is here: Point Cloud Library (PCL). 2011 IEEE International Conference on Robotics and Automation. 2011. P. 1–4.

Ruchay A., Gladkov A., Chelabiev R. Fast 3D object pose normalization for point cloud. Applications of Digital Image Processing XLIV. Vol. 11842. SPIE, 2021. DOI: 10.1117/12.2593893.

Ruchay A., Kalschikov V., Gridnev A., Guo H. Fast 3D object symmetry detection for point cloud. Applications of Digital Image Processing XLIV. Vol. 11842. SPIE, 2021. DOI: 10.1117/12.2593895.

Chollet F. et al. Keras. 2015. URL: https://github.com/fchollet/keras.

Tan M., Le Q. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning. Vol. 97 / ed. by K. Chaudhuri, R. Salakhutdinov. PMLR, 2019. P. 6105–6114. Proceedings of Machine Learning Research.




DOI: http://dx.doi.org/10.14529/cmse230101